A large number of molecular, cellular, and epidemiologic factors have been implicated in the regulation of bone development. A major unsolved problem is how to integrate these disparate findings into a concept that explains the development of bone as an organ. Often events on the organ level are simply presented as the cumulative effect of all factors that individually are known to influence bone development. In such a cumulative model it must be assumed that each bone cell carries the construction plan of the entire skeletal anatomy in its genes. This scenario is implausible, because it would require an astronomical amount of positional information. We therefore propose a functional model of bone development, which is based on Frost's mechanostat theory. In this model the genome only provides positional information for the basic outline of the skeleton as a cartilaginous template. Thereafter, bone cell action is coordinated by the mechanical requirements of the bone. When mechanical challenges exceed an acceptable level (the mechanostat set point), bone tissue is added at the location where it is mechanically necessary. The main mechanical challenges during growth result from increases in bone length and in muscle force. Hormones, nutrition, and environmental factors exert an effect on bone either directly by modifying the mechanostat system or indirectly by influencing longitudinal bone growth or muscle force. Predictions based on this model are in accordance with observations on prenatal, early postnatal, and pubertal bone development. We propose that future studies on bone development should address topics that can be derived from the mechanostat model.